Knowledge Commons of Institute of Automation,CAS
A Teacher-Student Framework for Maintainable Dialog Manager | |
Wang, Weikang1,2; Zhang, Jiajun1,2; Zhang, Han3; Hwang, Mei-Yuh3; Zong, Chengqing1,2,4; Li, Zhifei3 | |
2018-11 | |
会议名称 | 2018 Conference on Empirical Methods in Natural Language Processing (EMNLP) |
会议日期 | 2018-11 |
会议地点 | 布鲁塞尔 |
摘要 | Reinforcement learning (RL) is an attractive solution for task-oriented dialog systems. However, extending RL-based systems to handle new intents and slots requires a system redesign. The high maintenance cost makes it difficult to apply RL methods to practical systems on a large scale. To address this issue, we propose a practical teacherstudent framework to extend RL-based dialog systems without retraining from scratch. Specifically, the “student” is an extended dialog manager based on a new ontology, and the “teacher” is existing resources used for guiding the learning process of the “student”. By specifying constraints held in the new dialog manager, we transfer knowledge of the “teacher” to the “student” without additional resources. Experiments show that the performance of the extended system is comparable to the system trained from scratch. More importantly, the proposed framework makes no assumption about the unsupported intents and slots, which makes it possible to improve RL-based systems incrementally |
文献类型 | 会议论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/39123 |
专题 | 多模态人工智能系统全国重点实验室_自然语言处理 |
作者单位 | 1.National Laboratory of Pattern Recognition, Institute of Automation, CAS, Beijing, China 2.University of Chinese Academy of Sciences, Beijing, China 3.Mobvoi AI Lab, Seattle, US 4.CAS Center for Excellence in Brain Science and Intelligence Technology, Beijing, China |
第一作者单位 | 模式识别国家重点实验室 |
推荐引用方式 GB/T 7714 | Wang, Weikang,Zhang, Jiajun,Zhang, Han,et al. A Teacher-Student Framework for Maintainable Dialog Manager[C],2018. |
条目包含的文件 | 下载所有文件 | |||||
文件名称/大小 | 文献类型 | 版本类型 | 开放类型 | 使用许可 | ||
2018-WeikangWang-emn(1308KB) | 会议论文 | 开放获取 | CC BY-NC-SA | 浏览 下载 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论